P-value of expected/observed proportion
0
0
Entering edit mode
5.0 years ago

Statistics question (I formalise a real biological question to explain it easier):

There is an urne with 1000 balls - 600 white and 400 black. You randomly pick up 100 balls, and observe 40 white and 60 black ones. How to calculate the p-value of whether or not the observed proportion is significantly different from the random expecation?
Would a simple chi-square test do?

stats • 1.6k views
ADD COMMENT
0
Entering edit mode

ehm. I think there is something wrong with this formulation. You have an urn with 600/400 white/black balls and randomly picked up 100 balls. Your selected sample by definition is under null hypothesis (you took it from your initial urn). Why would you care about the p-value here?

ADD REPLY
1
Entering edit mode

Hi German, I want to see if there is a bias/selection towards picking-up, for example, white balls. BTW, congrats :)

ADD REPLY
1
Entering edit mode

Vielen Dank =) oh, so then it was not randomly selected? Use hypergeometric pdf https://en.m.wikipedia.org/wiki/Hypergeometric_distribution

ADD REPLY
1
Entering edit mode

De nada :) and thanks reaching!

ADD REPLY
0
Entering edit mode

This isn't a bioinformatics question, so you'll get better/more responses over at Cross Validated.

ADD REPLY
1
Entering edit mode

Doubt there will be better responses than mine!

ADD REPLY
2
Entering edit mode

There most definitely will be - it's the statistics SE.

ADD REPLY
0
Entering edit mode

You are actually right, I changed my answer to correct one

ADD REPLY

Login before adding your answer.

Traffic: 2144 users visited in the last hour
Help About
FAQ
Access RSS
API
Stats

Use of this site constitutes acceptance of our User Agreement and Privacy Policy.

Powered by the version 2.3.6